Electrocardiographic criteria for left ventricular hypertrophy (LVH) were examined in 84 unselected pediatric patients with rheumatic heart disease. There were 47 male and 37 female patients, ranging in age from 6 to 19 years. Electrocardiographic LVH was detected in 41 patients (48.8%), i.e. in 55.3% (26/47) of boys and in 36.6% (15/41) of girls. Echocardiographically determined LVH was present in 42 cases (50%) if left ventricular mass (LVM) was indexed for height, or 47 cases (56%) if LVM was indexed for body surface area (BSA). The overall sensitivity of height-indexed electrocardiographic diagnosis of LVH was 71.4% (95% confidence interval = 57.7% to 85.1%), while its sensitivity was 73.8% (95% confidence interval = 60.0% to 87.0%). For BSA-indexed echocardiographic LVH, the sensitivity was 68.1% (95% confidence interval = 54.8 to 81.4%) and the specificity was 75.7% (95% confidence interval = 61.9% to 89.5%). When sex-adjustment was examined, there was no increase of sensitivity of electrocardiographic LVH. Sensitivity of the electrocardiogram for LVH increased when age-adjustment was examined with 13 years of age as a cut-off point, both for height-indexed and BSA-indexed echocardiographic LVH. Reasons for the difference between these findings and the findings in adult patients (remarkably low sensitivity and very high specificity of ECG LVH) were discussed. Electrocardiogram was a moderate diagnostic modality in the detection of LVH in our pediatric patients with rheumatic rheumatic heart disease. Sex did not influence the sensitivity of ECG LVH, but older age group tended to increase its sensitivity.